{"credit":"Taylor (gtmcknight), github.com/gtmcknight/amm-sim@4812a6382311624e1506aca131ca35c378f930d9 research/solved; original replica, lattice/Gaussian filters, fee policy and tables. LEANBASE and its tell are credited adaptations.","idea":"Replace only LEANBASE price filtering with Taylor lattice Bayes; retain observable tell, rival pull, fees and qm tables. Measure truncation and spacing sensitivity.","rule":{"state_kept":"LEANBASE state plus moving log-price centre, cell probability masses and support bounds. No true price, rival spot, trade flag or run parameters reach the live policy.","per_callback":"LEANBASE WAD-observable tell sets the arb likelihood (negative=>retail; positive=>chance-scaled retail likelihood). Diffuse each elapsed step with Taylor Gaussian kernel, clip silent steps to no-arb band, mix point-mass arb and band-truncated retail posterior; moment-match into unchanged qm lookup. Radius arms zero cells outside +/-k indices from nearest cell to the current belief mean and renormalize after diffusion and posterior; full/double retain native moving-window recentering.","integration_details":["Taylor cell width=0.00025, 121 cells, sigma=0.000945; double uses width=0.0005 and 61 cells for same +/-150 bps extent.","Kernel radius ceil(4*sigma/width): 16 (33 taps), or 8 (17 taps).","Lattice native hard threshold 0.999 retained as gates uses it; LEANBASE Gaussian soft threshold is 0.995.","Silent cap kept at LEANBASE 12 rather than native lattice 30 to isolate the filter within existing callback work bound.","Rename lattice pin method to avoid collision with TellPolicy.pin boolean; dynamic real.RHO preserves tell chance multiplier.","Diffusion loop skips out-of-range taps with equivalent clipped range; seed 64000 full/double and k24/k32 scores unchanged after this optimization."]},"choose":{"seeds":[64000,64015],"n":16,"arms":[{"name":"full","oracle":false,"minus_base":{"mean_diff":0.003895349511903845,"se":0.03314389680382488,"min":-0.20389408629483796},"n":16},{"name":"k16","oracle":false,"minus_base":{"mean_diff":-0.28173904288606977,"se":0.0748092346387955,"min":-0.9629439279832468},"n":16},{"name":"k24","oracle":false,"minus_base":{"mean_diff":0.01069833141741583,"se":0.03181775159010102,"min":-0.17635199539734003},"n":16},{"name":"k32","oracle":false,"minus_base":{"mean_diff":0.009166361055527261,"se":0.03418196262021432,"min":-0.20393166559580322},"n":16},{"name":"double","oracle":false,"minus_base":{"mean_diff":0.011140055007853533,"se":0.030473682071936008,"min":-0.16080166251640549},"n":16}],"rejected":"k16: score.py stopped worse at n=16, mean -0.2817391875, SE 0.07480923756 (six-decimal score adapter). k4/k8 were single-seed exploratory timing probes, also poor.","selected_radii":[20,24,32],"selection_rule":"Keep full and doubled spacing; replace clearly losing k16 by intermediate k20; k24 and k32 retain wider tails. k20 has only one selection probe, seed 64000, edge 439.6286474051225 vs base 439.66988051382555; no claim it is tuned optimal.","probes":{"seed":64000,"k4_edge":412.1726688753176,"k8_edge":415.8065499963883},"source":"code_zb64:selection_evidence.json; selection probe outputs reproduced in this object"},"frozen":{"credit":"Taylor (gtmcknight), github.com/gtmcknight/amm-sim@4812a6382311624e1506aca131ca35c378f930d9 research/solved","params":{"arms":["base","full","k20","k24","k32","double"],"options":{"soft":1,"qm":"qm_k1e-5.json","a0":0.35,"ainf":0.35},"spacing":0.00025,"k":[20,24,32],"double_spacing":0.0005,"silent_cap":12,"hard":0.999,"report_plan":"256 seeds, 10000..10255, >2-second timing allowance; fixed before inspecting report results."},"code_sha256":"211ba490b39565984b2b508ece2afa3b94e4cf31c774feb27a812730e286baf8","manifest":{"experiment_lattice.py":"cce3fb928f6261a8ded847aa59a0e670436dcec3b4ab54c4c7d832493bc9674c","lattice_lean.py":"2d2717dc0ab1326cc9b14befe4491fb74f3ebc82bf3a601e4b71fcf1e40621d0","leanbase.py":"ba015769f7981d7798acbbc9aff8c0e0c59825bb1541d55bcdef49311d7a04bc","lean_tell.py":"b70054f8c0561779e40c4c8df3f41ab08da05b9aef591816ac6c9034beb40a95","stalerival.py":"d124a171d35b024daed840e5449d352b2b5e50a68ce98c8a9aaafaf3419d23da","tellpol.py":"7d5961c45bec56292533e65607245ad119b8733c760a5bf28f76a2e07bc3960b","tellref.py":"4867a33106ce2597863b8d7c8fb396288954b98f291714801fae8820f1f47aa3","real.py":"6a4a0f0e7f8b43d38e5eaa9d8b2e847a0771aee7f89551b122fe89bfe7969b5f","replica2.py":"54380c8755f4e42e9a0426759e918eacd2c5f915b7f04e5d0501c394c5846b40","run.py":"91133e4ef485626226679d73349b5b4e374d9713ac5d7e9d36a39e1afd75933c","par.py":"bf8a1a2e2a4bf9a23bf4622d726fd815702298ad1f9fa0dd17197230cce63ec1","evalpol.py":"79a96efe8c543aee125c937554688b37ec2e7df44df4f3435265bba6eb012bf3","pol_mid.json":"efbb5ee102db4d312f64565d9bf7232b6becbe02b624164f744a402bbbf2983b","qm_k1e-5.json":"5ff53e42ca87390e5f65d1caa3f7a480e67a8212fb0670274dfd562b05ba1afd"},"time":1794332574.4089248,"table_urls":{"pol_mid.json":"https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/solved/pol_mid.json","qm_k1e-5.json":"https://raw.githubusercontent.com/gtmcknight/amm-sim/66604b3426d8e97a73e2c39940b9b4dece51cc73/research/solved/qm_k1e-5.json"},"code_sha256_definition":"SHA256 of canonical JSON path:text map of the .py files in frozen.manifest, sort_keys=True,separators=(comma,colon); evidence files added to bundle after freeze."},"report":{"seeds":[10000,10255],"n":256,"arms":[{"name":"full","oracle":false,"minus_base":{"mean_diff":0.010143564101050995,"se":0.009039630931854808,"min":-0.3250121475361425},"n":256,"nodes_per_step":{"mean_active_input":15.154466845951669,"max_active_input":76},"minus_full":{"mean_diff":0.0,"se":0.0,"min":0.0},"nominal_retained_nodes":121,"spacing_bps":2.5,"gain_fraction_of_full":1.0,"source":"code_zb64:paired_evidence.json; experiment_lattice.py; lattice_lean.py"},{"name":"k20","oracle":false,"minus_base":{"mean_diff":-0.024020486358576232,"se":0.01242757362187408,"min":-0.48449584380364286},"n":256,"nodes_per_step":{"mean_active_input":14.60223972952558,"max_active_input":41},"minus_full":{"mean_diff":-0.034164050459627227,"se":0.009691483214660597,"min":-0.6064255907157303},"nominal_retained_nodes":41,"spacing_bps":2.5,"gain_fraction_of_full":-2.3680519114664462,"source":"code_zb64:paired_evidence.json; experiment_lattice.py; lattice_lean.py"},{"name":"k24","oracle":false,"minus_base":{"mean_diff":-0.0020268365647975184,"se":0.010316582124862729,"min":-0.4547366756984843},"n":256,"nodes_per_step":{"mean_active_input":14.874847502241062,"max_active_input":45},"minus_full":{"mean_diff":-0.012170400665848513,"se":0.005742371100386745,"min":-0.507545422275598},"nominal_retained_nodes":49,"spacing_bps":2.5,"gain_fraction_of_full":-0.19981502996441988,"source":"code_zb64:paired_evidence.json; experiment_lattice.py; lattice_lean.py"},{"name":"k32","oracle":false,"minus_base":{"mean_diff":0.01165086180457453,"se":0.008859272427411347,"min":-0.3247362804833642},"n":256,"nodes_per_step":{"mean_active_input":15.085068154091317,"max_active_input":54},"minus_full":{"mean_diff":0.0015072977035235358,"se":0.00167632949811607,"min":-0.12449454070281263},"nominal_retained_nodes":65,"spacing_bps":2.5,"gain_fraction_of_full":1.1485964586517832,"source":"code_zb64:paired_evidence.json; experiment_lattice.py; lattice_lean.py"},{"name":"double","oracle":false,"minus_base":{"mean_diff":0.012092851766208446,"se":0.009886798901639855,"min":-0.32446977624601914},"n":256,"nodes_per_step":{"mean_active_input":8.076311071443644,"max_active_input":39},"minus_full":{"mean_diff":0.0019492876651574509,"se":0.006595829021622125,"min":-0.45502015811825913},"nominal_retained_nodes":61,"spacing_bps":5.0,"gain_fraction_of_full":1.192169896669306,"source":"code_zb64:paired_evidence.json; experiment_lattice.py; lattice_lean.py"}],"order":"Consecutive 64-seed blocks in order, every arm in each block before the next. Each block is split into two 32-seed commands to stay below 8 minutes; no report seed is rerun.","rival":"Replica constant 30 bps; each policy run independently and paired by seed against LEANBASE.","uncertainty":"Sample standard deviation of paired per-seed edge differences divided by sqrt(n); min is worst paired seed.","smallest_lattice_retaining_half_full_gain":{"name":"double","nominal_nodes":61,"threshold":0.005071782050525497,"qualification":"Descriptive comparison of means, not a certified half-gain guarantee; null if full has no positive gain."},"minimum_sample_rule":"Full lattice took 2.5477 seconds on initial 10000-step selection seed, permitting the 256-seed minimum.","nodes_definition":"Nonzero input masses >1e-15 per diffuse call, including silent steps; executed kernel work, not array capacity.","block_columns":["mean_diff","se","min"],"blocks":[{"seeds":[10000,10063],"arms":{"full":[-0.01019891018448682,0.01705496972977634,-0.2648283937691076],"k20":[-0.040187254451024224,0.019923370648575106,-0.48449584380364286],"k24":[0.0013910286467000432,0.02423733869288049,-0.4028907176518146],"k32":[-0.009886899426438944,0.01691887220474777,-0.25472435211975153],"double":[0.0002770637789444308,0.018188423188173142,-0.2980319696995366]}},{"seeds":[10064,10127],"arms":{"full":[0.021970215566572193,0.019991540583459283,-0.3120671702848199],"k20":[-0.01191569305239959,0.021039667098398256,-0.33015126463874367],"k24":[0.005928125154467168,0.019668098398704342,-0.2786235116345779],"k32":[0.028813875762635277,0.019797699991327376,-0.28955159608011627],"double":[0.019662946044515905,0.018816242329455047,-0.2840210768506495]}},{"seeds":[10128,10191],"arms":{"full":[0.025800073083734,0.017037009012075402,-0.3250121475361425],"k20":[-0.004961538390154452,0.0348961346341574,-0.4334829123644113],"k24":[0.002573554269408973,0.019264113015953564,-0.31392357081028877],"k32":[0.02562509272586322,0.01684259645138564,-0.3247362804833642],"double":[0.040472360214762304,0.025119526417124106,-0.3086626152233407]}},{"seeds":[10192,10255],"arms":{"full":[0.003002877938384607,0.018130673131389295,-0.29667308310047247],"k20":[-0.03901745954072666,0.02067299408081909,-0.4423565354999255],"k24":[-0.018000054329766257,0.01929295597609937,-0.4547366756984843],"k32":[0.0020513781562385702,0.01709578292806815,-0.2966711359189276],"double":[-0.012040962973388858,0.015542982629745586,-0.32446977624601914]}}],"evidence_encoding":"code_zb64 contains paired_evidence.json: contiguous seeds from seed_start; n rows; columns name six independent policy edges; edges_f64_le_b64 is base64 of row-major little-endian IEEE-754 float64 values, preserving exact simulator outputs. read_evidence.py decodes it with stdlib."},"cost":{"extra_storage_words":{"full":123,"k20":44,"k24":52,"k32":68,"double":63},"operation_unit":"Convolution multiply-accumulate (fixed-point MUL, DIV, ADD); excludes O(N) clipping, moments, normalization, trimming.","worst_ops_per_callback":{"full":51909,"k20":18018,"k24":21450,"k32":28314,"double":13481},"ops_per_silent_step":{"full":3993,"k20":1386,"k24":1650,"k32":2178,"double":1037},"gas_estimate":{"full":{"convolution_only_optimistic":674817,"measured":null},"k20":{"convolution_only_optimistic":234234,"measured":null},"k24":{"convolution_only_optimistic":278850,"measured":null},"k32":{"convolution_only_optimistic":368082,"measured":null},"double":{"convolution_only_optimistic":175253,"measured":null}},"how_estimated":"Direct dense convolution upper-work bound: 12 silent diffusions plus current diffusion, N input masses times 33 taps (double:17). Reduced arms allow one extra retained cell for native linear recentering. Optimistic fixed-point MAC cost assumed 13 gas = MUL 5 + DIV 5 + ADD 3; excludes loop, memory, clipping, nonlinear functions, storage, table and tell. This is an engineering estimate, not compiled gas or a lower bound for redesigned algorithms. Extra storage estimates a ring-buffer port with one word per retained mass plus centre and packed bounds (Python still allocates the full array); not an implemented contract. No evidence that any direct arm fits 40000 gas."},"advance":false,"reading":["Against LEANBASE on 256 paired replica seeds, full lattice changes edge by +0.010144 (SE 0.009040), so the three-seed gains against Taylor\u2019s original Gaussian policy do not by themselves establish a gain over this stronger baseline.","Radii 20, 24 and 32 change edge by -0.024020, -0.002027 and +0.011651, while double spacing gives +0.012093; the smallest half-gain arm is double, using means rather than a confidence guarantee.","No arm reaches the required +0.02 mean with positive mean minus two SE, and direct convolution also lacks a demonstrated 40000-gas worst-callback implementation."],"code_zb64":"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","seeds_used":{"selection":[64000,64015],"timing_and_parity_probes":[64000],"report":[10000,10255],"engine":[]},"not_done":["No Solidity port, engine trials, measured gas or submission: this is the requested Python-only research item.","Did not run Taylor gates on 10000..10002 before freezing, to avoid report-seed leakage; gate source inspected only.","Full 2048-seed report not attempted within the 40-minute work window; use the >2-second minimum-sample allowance.","Large immutable pol_mid.json and qm_k1e-5.json tables are not embedded in code_zb64; their pinned hashes are in frozen.manifest and URLs in sources.json. No packages installed; runner uses Python standard library."],"validation":{"filter_unit_checks":"Passed all five arms: kernel moments, clipping, normalization, support bounds, pin; baseline dispatch restored.","reference_integrity":"LEANBASE, lean_tell, stalerival, tellpol, tellref, real and replica2 SHA256 match the supplied BASE artifact.","source_quotes":"Five exact public-source quotations fetched and verified locally."}}